guyfawkes wrote:My definition of a successful hobby OS
Whatever it is, it's right for you, so everyone can make their own valid list. The same applies to the OP. If you want to average out everyone's ideas of what's successful, you need to do some polling, and that'll give you a better idea of what people generally think. But does it really count for anything? I'll just go through your list (which isn't bad, by the way) as it's to hand:-
1. ***It needs to be self hosting (you can compile/assemble and code the OS, on the OS its self.
That can't be essential.
2. It needs some form of basic network
That isn't essential either.
3. ***It needs to be able to load and run app
Unless it's only been designed to illustrate some aspect of OS design.
3. It needs both a text and graphical interface
Unless it's an OS for blind users.
4. ***It's code must be 100% original
That's going to rule out just about any OS you can find, apart from the original one.
5. ***It must have been available to test for at least 5 years.
Why? A brand new one could be far better and become much more widely used within its first year.
6. ***It must have been independently chosen as a good example of a hobby OS.
Chosen by who? One of the small group discussing the issue?
The only way for an OS to qualify as successful is for it to succeed at doing something. However, the original question is about being more successful, so what you actually need to do is look at how many things the OS attempts to do and how many it succeeds in doing, plus how well it does them - the more successful ones will try to do more and succeed in doing more of those things well. There's then another aspect of being successful, and that's actually being used by people: two OSes may be functionally identical, but one is used widely while the other isn't. This is largely down to luck (getting in first, better promotion, or simply having more apps available). So, what you really want to do is create two lists of OSes, one representing the number of users (or better still the total amount of time they've been running, or better still the amount of useful work they've done), and the other list would rank them in order of actual capability combined with how well they work. Even then, the distinction between hobby OS and non-hobby OS is arbitrary, so you've got a problem deciding where to draw the line, and the more successful projects may reach the point where they get too successful and no longer qualify, so why make such a pointless distinction in the first place?
One thing's for sure - if you want your OS to become one of the successful ones, you want to avoid wasting any more of your time discussing how long a piece of string is.
.
.
